Fifteen-plus years of BC mortgage experience, distilled into practical writing on buying, renewing, refinancing, and building wealth through real estate.
CIBC Says StatCan's Population Count Is Wrong by Hundreds of Thousands
Statistics Canada is about to overhaul how it counts non-permanent residents, and the revision could erase the population decline that made headlines earlier this year. CIBC Capital Markets estimates the adjustment will add hundreds of thousands of people to the official tally, people who were already here, just not showing up in the numbers.
The gap stems from how expired visas get tracked. When a work permit or study permit lapses, StatCan's quarterly estimates assume the holder left the country unless they applied for renewal. But many didn't leave. They stayed under implied status, a legal limbo where someone can remain in Canada while their application processes. Others overstayed without documentation. The current methodology treats both groups as departed, even when they're still competing for rental units and showing up to emergency rooms.
Benjamin Tal, CIBC's deputy chief economist, flagged this as a structural miscount, not a rounding error. The bank's analysis suggests the real 2026 population baseline could be several hundred thousand higher than what StatCan published in the first quarter. That means the "cooling" narrative, the idea that temporary resident numbers were finally dropping after years of rapid growth, was partly statistical artifact. The people didn't leave. They just fell through the counting system.
Why the denominator matters
If the population is larger than reported, every per-capita figure shifts downward. GDP per capita, already negative for multiple quarters in 2025 and early 2026, gets worse when you divide by a bigger denominator. Productivity metrics worsen. The math is clean: same output, more people, lower average prosperity.
Housing projections break the other direction. The Canada Mortgage and Housing Corporation estimates the country needs 3.5 million additional units by 2030 to restore affordability. If the baseline population in 2026 is understated by hundreds of thousands, latent demand is higher than CMHC modelled. Developers and municipalities planning for a slowdown in household formation may be building into a tighter market than they realized.
Federal transfer payments also run off these numbers. Provinces receive healthcare and infrastructure funding partly based on official headcounts. An upward revision could mean Ontario, British Columbia, and Quebec have been underfunded relative to the actual service load their systems are carrying.
Policy designed for ghost data
The federal government set a target in 2024 to bring non-permanent residents down to 5% of the total population by 2027. The caps on student visas and temporary foreign worker permits rolled out in 2025 and early 2026 were calibrated to hit that figure. But if the starting point was wrong, if the NPR share was already higher than reported, the caps aren't aggressive enough to meet the stated goal. You can't design an effective throttle when the speedometer is broken.
Immigration, Refugees and Citizenship Canada relies on the same quarterly demographic estimates everyone else does. The disconnect between arrivals data (which IRCC controls) and population-in-country estimates (which StatCan models) has been widening since 2022, when temporary migration surged faster than the tracking systems could handle.
What gets revised and when
StatCan runs a post-Census reconciliation every few years to correct over-coverage and under-coverage in its quarterly models. The 2026 cycle, expected later this year, will integrate findings from the 2021 Census with updated administrative data on renewals, departures, and implied-status cases. The agency has acknowledged gaps in how it models NPR exits and signalled that the upcoming revision will address them.
The correction won't mean new people arrived overnight. It means they were here all along, using services, renting apartments, working off-books or on extended permits, just not appearing in the official count. The revision converts an administrative invisibility into a recorded fact. That fact has consequences for every policy lever tied to population: housing targets, healthcare capacity, labour market forecasting, central bank rate decisions.
The quarters of "decline" may not have been decline at all. Just a tracking failure catching up with reality.
Statistics Canada is about to overhaul how it counts non-permanent residents, and the revision could erase the population decline that made headlines earlier this year. CIBC Capital Markets estimates the adjustment will add hundreds of thousands of people to the official tally, people who were already here, just not showing up in the numbers.
The gap stems from how expired visas get tracked. When a work permit or study permit lapses, StatCan's quarterly estimates assume the holder left the country unless they applied for renewal. But many didn't leave. They stayed under implied status, a legal limbo where someone can remain in Canada while their application processes. Others overstayed without documentation. The current methodology treats both groups as departed, even when they're still competing for rental units and showing up to emergency rooms.
Benjamin Tal, CIBC's deputy chief economist, flagged this as a structural miscount, not a rounding error. The bank's analysis suggests the real 2026 population baseline could be several hundred thousand higher than what StatCan published in the first quarter. That means the "cooling" narrative, the idea that temporary resident numbers were finally dropping after years of rapid growth, was partly statistical artifact. The people didn't leave. They just fell through the counting system.
Why the denominator matters
If the population is larger than reported, every per-capita figure shifts downward. GDP per capita, already negative for multiple quarters in 2025 and early 2026, gets worse when you divide by a bigger denominator. Productivity metrics worsen. The math is clean: same output, more people, lower average prosperity.
Housing projections break the other direction. The Canada Mortgage and Housing Corporation estimates the country needs 3.5 million additional units by 2030 to restore affordability. If the baseline population in 2026 is understated by hundreds of thousands, latent demand is higher than CMHC modelled. Developers and municipalities planning for a slowdown in household formation may be building into a tighter market than they realized.
Federal transfer payments also run off these numbers. Provinces receive healthcare and infrastructure funding partly based on official headcounts. An upward revision could mean Ontario, British Columbia, and Quebec have been underfunded relative to the actual service load their systems are carrying.
Policy designed for ghost data
The federal government set a target in 2024 to bring non-permanent residents down to 5% of the total population by 2027. The caps on student visas and temporary foreign worker permits rolled out in 2025 and early 2026 were calibrated to hit that figure. But if the starting point was wrong, if the NPR share was already higher than reported, the caps aren't aggressive enough to meet the stated goal. You can't design an effective throttle when the speedometer is broken.
Immigration, Refugees and Citizenship Canada relies on the same quarterly demographic estimates everyone else does. The disconnect between arrivals data (which IRCC controls) and population-in-country estimates (which StatCan models) has been widening since 2022, when temporary migration surged faster than the tracking systems could handle.
What gets revised and when
StatCan runs a post-Census reconciliation every few years to correct over-coverage and under-coverage in its quarterly models. The 2026 cycle, expected later this year, will integrate findings from the 2021 Census with updated administrative data on renewals, departures, and implied-status cases. The agency has acknowledged gaps in how it models NPR exits and signalled that the upcoming revision will address them.
The correction won't mean new people arrived overnight. It means they were here all along, using services, renting apartments, working off-books or on extended permits, just not appearing in the official count. The revision converts an administrative invisibility into a recorded fact. That fact has consequences for every policy lever tied to population: housing targets, healthcare capacity, labour market forecasting, central bank rate decisions.
The quarters of "decline" may not have been decline at all. Just a tracking failure catching up with reality.
Read Next
Brokers Who Think the Filogix Acquisition Changes Nothing Are Missing the Strategic Shift
DLC didn't buy Filogix to own the rails, it bought it because the rails were already failing
A Toronto Townhouse Project Built 147 Units and Can't Sell 65 of Them
MCAN's 19% earnings jump hides a rising impairment problem